Hallo everyone.
The very simple SPLUS example I would like to present is the following:
I'm dealing with a table X of length 5.
> X
2 8 10 16 24
105 63 30 46 1
> names(X)
[1] "2" "8" "10" "16" "24"
> X["2"]
2
105
I want now to extract the table position with the name "1" (which of
course doesn't exists)
> X["1"]
Problem in X["1"]: Array subscript (6) out of bounds, should be at most 5
Use traceback() to see the call stack
The convertion of the table X in a data.frame Y yields:
> Y<-data.frame(X)
> Y
X
2 105
8 63
10 30
16 46
24 1
and
> Y["1",]
[1] NA
> version
Version 6.0 Release 1 for Linux 2.2.12 : 2000
I computed the same example on the same machine with R. The results are
listed below.
> X
2 8 10 16 24
105 63 30 46 1
> names(X)
[1] "2" "8" "10" "16" "24"
> X["2"]
2
105
> X["1"]
NA
NA
> Y<-data.frame(X)
> Y
X
2 105
8 63
10 30
16 46
24 1
> Y["1",]
NA
NA
> version
platform i686-pc-linux-gnu
arch i686
os linux-gnu
system i686, linux-gnu
status
major 1
minor 3.1
year 2001
month 08
day 31
language R
My aim is not to compare the performances of Splus and R. But, according
to this simple example, I suppose some lack of consistency in SPLUS
between table and data.frame.
Regards.
Dr. Pierre Ilouga
EVOTEC OAI
Screening Operations/Discovery Informatics
Schnackenburgallee 114
D-22525 Hamburg
Germany
|